Zhijing Yu

Precision Circular Target Location in Vision Coordinate Measurement System

The precision circle target sub-pixel centroid location is one of the key factors of large space 3D coordinates measurement. In this paper, the circle retro-reflective targets and optical targets used in large space 3D coordinates measurement are analyzed, and the characteristics and applied fields of target sub-pixel centroid location algorithms such as gray weighted and squared gray weighted centroid location algorithms which combine with bi-linear interpolation or improved cubic convolution interpolation method, are discussed. The influencing factors and improving methods for circle target location are also analyzed. These factors include the resolution of the CCD camera, image grab noise, the size of targets, the choice of exposure time and threshold value, the noncoincidence of centroid and center of the ellipse imaging targets, the distortion influence of deformational targets and so on. Based on these, proper compensating measures are brought out for the problems such as target choice and location, influencing factors and eliminating measures.

Coded Target Based Traffic Accident Vision Swift Reconnaissance System

Aimed at the defects of current traffic accident rapid investigation methods, a traffic accident vision swift reconnaissance system based on coded target is proposed. Make use of the uniqueness and feature stations geometric constraint of coded target, and we can realize automatic recognition and match of coded target and precise space scale constraint. The coded targets distributed in the accident scene are imaged by the CCD camera in several locations and orientations, and the 3D coordinates of space measured feature points are automatically solved through the initial external location and orientation of camera based on co-angular constraint pose estimate algorithm and full-scale bundle adjustment algorithm based on collinear constraint. Therefore the 3D information of the traffic accident scene is swiftly acquired and the traffic accident scene is also reconstructed. The experiment results demonstrate that the proposed method is feasible and flexible.

Research on Camera Internal Parameters Self-Calibration Method for Mobile 3D Coordinates Vision Measurement System

According to the requirement of characteristics of large-scale 3D coordinates on-line measurement system, a method of mobile 3D coordinates measurement camera internal parameters self-calibration was presented in this paper. Through measurements of multiple stations, sufficient pictures of coded targets and un-coded ones were obtained. And a bundle adjustment calibration method based on collinear constraint in computer vision field was used. With the diverse initial values, different calibration results were analyzed to compare the precision. And the experiment results showed the method we presented could obtain preferable results for large-scale 3D coordinates measure occasions. KeywordsSelf-calibration; Camera internal parameters; mobile vision; 3D coordinates measurement

Application of improved cubic convolution interpolation in circular target location

The precision of target sub-pixel centroid location directly affects the result of large scale vision 3D coordinates measurement. This paper deeply studies the sub-pixel centroid location algorithm of retro-reflective targets and infrared optical targets used in 3D coordinates measurement system, and makes use of improved cubic convolution interpolation algorithm to increase the number of effective pixels used in centroid location, then gives optimizing adjustment parameters for different types of targets and combined with squared gray weighted centroid location algorithm, finally realizes accurate target sub-pixel centroid location. This algorithm is proved to be effective and robust by simulations and experiments.

Camera Calibration Method Used in Computer Vision Coordinate Measurement Systems

As one of the key techniques of vision coordinates measurement, camera internal parameters calibration is regarded by technologists and has been researched a lot. Based on the study of regular internal parameters calibration methods such as plane calibration, two-plane calibration, vanishing points calibration, stereo calibration and so on, combined with the requirement of large space vision coordinates measurement for camera internal parameters calibration, this paper brings forward independent internal parameters calibration and virtual stereo calibration methods. A special calibration equipment is used in independent internal parameters calibration to calibrate original points coordinates of camera and distortion parameters of lens, and a virtual target board and bundle adjustment algorithm are used in virtual stereo calibration to calibrate internal parameters. They can be used in large space vision coordinates measurement and get high calibration accuracy.